Abstract 3689: Genome Wide Association Study of In-stent Restenosis Identifies Multiple Candidate Susceptibility Loci
Introduction: Risk-conferring genes responsible for complex diseases are characterized by variable expressivity, low penetrance, epistasis and locus heterogeneity, making the analysis of complex genetic traits challenging. For complex cardiovascular diseases, candidate gene approaches have achieved only limited success, making unbiased genome-wide association study (GWAS) designs appealing. Molecular and genetic studies suggest that in-stent restenosis (ISR) is primarily an inflammatory and proliferative disease, with distinct roles for cell cycle proteins, growth factors, and inflammatory cytokines.
Methods: To investigate the genetic basis of ISR, we designed a case control GWAS on a high-density dataset of ~116,000 single nucleotide polymorphisms (SNPs) assayed in 407 patients (150 cases, 257 controls). We undertook a haplotype analysis of regions highlighted by the presence of two or more SNPs within 250 kb of one another and with p<0.001 (unadjusted) in univariate tests of allelic association. Haplotypes were defined in the regions and tested for association with ISR, using a global Bonferroni correction for the 84 haplotypes tested.
Results: We identified eight candidate susceptibility loci for ISR, five of which contained the genes NOV, ARNTL, TAF4B, PKP4 and a hypothetical protein FLJ21986. We observed expression of these genes by RNA and protein analysis in normal, atherosclerotic and restenotic human coronary arteries, supporting the hypothesis that these genes may mediate vascular homeostasis and adverse vascular remodeling. We estimated each subject’s haplotypes and formed two haplotype clusters. To further define the relationship of these haplotypes with the ISR phenotype, we examined the time to development of ISR as a quantitative trait indicating disease severity in a Cox regression model and demonstrated a significant effect of allele dose for each of the eight regions identified (p-value 0.00038 to 0.035, after Bonferroni correction), suggesting that allele copy number contributes to the development of ISR.
Conclusions: Haplotype analysis of GWAS using SNP markers is a useful approach to identify candidate genes for chronic ischemic heart diseases. These areas of association with ISR warrant further investigation.